Click here to show/hide codes in this notebook.

About the data¶

  • The data was collected and made available by Dr. Kristen Gorman and the Palmer Station, Antarctica LTER, a member of the Long Term Ecological Research Network.
  • It contains data for 344 penguins. There are 3 different species of penguins in this dataset, collected from 3 islands in the Palmer Archipelago, Antarctica.

Analysis Goal¶

  • As our dataset consists of penguins belonging to 3 different species, our goal of this analaysis is to produce methods / model to classify penguins into 3 different species.

Overview¶

Categorical Features Numerical Features
Species bill_length_mm
Sex bill_depth_mm
Island flipper_length_mm
body_mass_g

Understanding features

MarineGEO circle logo

Missing values in total and per species¶

  • Given sample contains maximum missing values in sex feature as compared to others. There is 5.5% of data missing and can be ignored or imputed based on other variables
  • Chinstarp has 0 missing values whereas gentoo and adelie do have missing values. Adelie contributes to nearly 3% missing values and gentoo contributes 2.5% of the missing values

Species distribution¶

  • Sample consists of of uneven species distribution with more number of adelie followed by chinstrap and then gentoo

Species distribution on each Island¶

  • Adelie is the most common species that can be found in different islands
  • Gentoo species are mainly found on Biscoe island and Chinstrap on Dream island.

Gender distribution among species¶

  • Sample consists of equal distribtuion of male and female among adelie and chinstarp species whereas in gentoo, females are dominated by males
  • Gentoo penguins have higher body mass, flipper length and bill length, but lower bill depth
  • Adelie and Chinstrap having similar body mass and bill depth but demonstrate very different bill length. On an average , a Chinstrap penguin's bill length will be 10mm longer than one of an Adelie's

Why dont the penguins fly?¶

  • Data reveales that penguins have higher body mass and shorter flipper length than most flight birds.
  • Penguins have more body fat to prevent heat loss on land and in the sea while their powerful flippers make them very accomplished swimmers
  • Based on the above mentioned facts, Gentoo species have higher insulation and are great swimmers as compared to Adelie and Chinstrap which have similar body mass and flipper length
  • Generally, bills tend to be longer and thin in species that are primarily fis eaters and shorter and stouher in those that mainly eat krill
  • Here, Gentoo and Chinstrap have similar bills as compared to Adelie

Can penguins gender be distinguished based on their features?¶

  • Across all species, females consistently demonstrate lower body mass, bill length, bill depth and flipper length as compared to males

Feature Importance¶

  • Flipper length, bill_length and bill_depth contribute the most to identifying species

Conclusion¶

**Adelie** : "I have small flippers, short and wide beak and i am slim" .
**Gentoo** : "I have large flippers, long and thin beak and i am fatty" .
**Chinstrap** : "I have small flipper, long and wide beak and i am slim"
.

MarineGEO circle logo

  • Gentoo penguins are the largest of the 3
    • Avg Body Mass: 5.076 Kgs
    • Avg flipper length: 217mm
    • Avg bill measurements: 47.5mm (length) * 15mm (depth)
  • Chinstrap penguins are 2nd in comparison
    • Avg Body Mass: 3.73Kgs
    • Avg flipper length: 196mm
    • Avg bill measurements: 49mm (length) * 18mm (depth)
  • Adelie penguins are smallest of the 3
    • Avg Body Mass: 3.7Kgs
    • Avg flipper length: 190mm
    • Avg bill measurements: 39mm (length) * 18mm (depth)
Click here to show/hide codes in this notebook.